An integrated genomic and transcriptomic survey of mucormycosis-causing fungi

Abstract

Mucormycosis is a life-threatening infection caused by Mucorales fungi. Here we sequence 30 fungal genomes, and perform transcriptomics with three representative Rhizopus and Mucor strains and with human airway epithelial cells during fungal invasion, to reveal key host and fungal determinants contributing to pathogenesis. Analysis of the host transcriptional response to Mucorales reveals platelet-derived growth factor receptor B (PDGFRB) signaling as part of a core response to divergent pathogenic fungi; inhibition of PDGFRB reduces Mucorales-induced damage to host cells. The unique presence of CotH invasins in all invasive Mucorales, and the correlation between CotH gene copy number and clinical prevalence, are consistent with an important role for these proteins in mucormycosis pathogenesis. Our work provides insight into the evolution of this medically and economically important group of fungi, and identifies several molecular pathways that might be exploited as potential therapeutic targets.

Figure 1. Phylogenetic relationships among 38 Mucorales fungi with three Entomopthorales as outgroups.

A phylogenetic tree was constructed using concatenated trimmed multiple alignments of 76 orthologous proteins shared among all 41 fungal strains. Bootstrap values (out of 100) are displayed at tree nodes. Of note, the R. oryzae and R. delemar sub-clade contains low bootstrap values throughout. All strains are clinical isolates unless otherwise indicated by a black circle to be environmental.

Figure 2. Presence of CotH invasins across Mucorales.

(a) HMM logo depicting conserved invasin domain. (b) Presence of CotH3 domain containing proteins in all 41 Mucorales fungi. Values represent –log of the e value from an HMM search of each positive hit (maximum e value of 1 × 10⁻⁴) versus the motif from a. Grey colour indicates no gene. (c,d) Relative gene expression levels (expressed as percentile rank RPKM for all of the genes in the genome) for the six CotH genes of R. delemar 99-880 (c) and R. oryzae 99-892 (d) during in vitro infection conditions. Blue and red bars indicate spores grown alone in tissue culture media for 6 h (blue) or 16 h (red). Green and purple bars indicate spores grown exposed to airway epithelial cells for 6 h (green) or 16 h (purple). Results are the mean±s.d. *P<0.05 versus 6-h sample in the same condition; n=3. Gene of the same name in R. delemar 99-880 and R. oryzae 99-892 represent 1-to-1 homologues.

Figure 3. Population structure of R. oryzae and R. delemar strains.

(a) SNP-based whole-genome maximum-likelihood phylogeny of all 13 R. delemar and R. oryzae strains. All nodes have a bootstrap value of ≥97 out of 100. (b) Population structure inferred using the program Admixture. Values represent fraction of population ancestry denoted by colours: green (R. delemar) and blue (R. oryzae).

Figure 4. Identification of orthologous clusters.

A bar plot of orthologous gene clusters in a subset of 19 Mucorales fungi. Core genes found in all 17 Rhizopus and Mucor strains are shown in blue, R. delemar/oryzae-specific genes are shown in red. Please refer to key for other classifications.

Figure 5. Host response to Mucorales fungi.

(a) The number of host genes that were induced (yellow) or repressed (blue) by infection with each of the Mucorales fungi at 6 or 16 h after addition of fungal spores. (b) Host upstream regulators that are predicted to be modulated by Mucorales infection of airway epithelial cells based on expression of known downstream targets. Red indicates predicted activation (z score>2). Blue indicates predicted repression (z score<−2). White indicates no predicted regulation. (c) The effect of pharmacological inhibition of the PDFG receptor on R. delemar 99-880-induced cell damage of endothelial cells. Results are the median±interquartile range of two experiments, each performed in triplicate. *P<0.018 versus control; n=6.